Children with Cerebral Palsy (CP) often exhibit crouch gait, which is characterized by excessive knee flexion during stance. This form of walking is extremely fatiguing and tends to progress until eventually walking ability ceases. A newly revived surgical procedure, distal femoral extension osteotomy (DFEO) and patellar tendon advancement (PTA), simultaneously addresses both knee flexion contractures and patella alta (superiorly displaced patella) that often co-exist in those with crouch. Initial outcoe studies demonstrate greater improvements in gait than conventional surgical treatments. However, complication rates remain high and some patients exhibit little to no improvement in gait after surgery. The first aim of this study is to investigate the effects of patella position, crouch and surgical parameters on functional knee mechanics. Computational knee models will be created that include detailed representations of ligament and articular cartilage geometry within the tibiofemoral and patellofemoral joints. Patella alta and knee flexion contractures will be introduced, and surgical simulations will be performed virtually. The computational models will then be used to simulate knee mechanics when walking in normal, and mild, moderate, and severe crouch gait postures. Probabilistic simulations will then be used to investigate how variability in physical characteristics and surgical factors can contribute to variable outcomes. The effects of surgery and crouch gait on cartilage pressure patterns will also be determined, which is relevant for understanding subsequent skeletal growth and long-term cartilage health. The second aim investigates whether a combination of quantitative measures of physical characteristics and surgical parameters can retrospectively classify post-surgical gait performance. Pre- and post-surgical x- rays will be used to quantitatively measure the Koshino index (metric of patella alta/baja), the magnitude and location of the DFEO, and the PTA advancement distance in patients who previously underwent DFEO+PTA. Pre-surgical measures of knee flexion contracture and spasticity will also be obtained. Pre- and post-operative gait analysis data will be used to assess changes in gait mechanics, while functional surveys will assess changes in performance on activities of daily living. The random forest algorithm will then be used to identify decision trees and associate predictor variables that can classify those patients whose gait and overall function improved after surgery. These results will be interpreted in the context of modeling results from Aim 1, thereby providing a potential mechanistic explanation for clinical observations. The research will impact innumerable CP patients as the DFEO+PTA surgical procedures are adopted around the world, and will also set the groundwork for more rigorous scientific study of other procedures used to treat gait disorders.